MAGEC3 is a prognostic biomarker in ovarian and kidney cancers

MAGEC3 protein normal tissue expression and cancer tissue expression

In normal tissues, MAGEC3 was widely expressed at a similar level except for decreased expression in breast and skin tissues and increased expression in tonsil, brain, and testis tissues (Figure 1A). Normal ovary tissues had a mean expression of 181.2 on a 0-300 point scale (H-score, median 176.5, sd 40.8) while unmatched ovarian cancers had significantly lower expression (mean 145.2, median 135.4, sd 40.4; two-sample t-test, p < 0.001) (Figure 1B). Normal kidney tissues had a similar mean expression of 174.9 (sd 20.9) and median expression of 171.6 (Figure 1D); non-tumor kidney tissue (not shown) in patients with renal cell carcinoma (RCC) was slightly lower (mean 167.8, median 170.7; two-sample t-test, p=0.003).

• Download figure
• Open in new tab

Figure 1. MAGEC3 protein expression in normal and cancer tissues.

(A) MAGEC3 protein expression in normal tissue cores. (B) MAGEC3 expression in normal ovary tissue and ovarian cancer (left side). Ovarian cancer expression levels (light green box) were dichotomized at the median into “Normal” (red) and “Loss” (blue) (right side). (C) Kaplan-Meier plot showing progression-free survival trends for ovarian cancers with optimal cytoreduction stratified at the cancer median. (D) MAGEC3 expression in normal kidney tissue and kidney cancer. Kidney cancer expression levels (light yellow box) were dichotomized at the median into “Normal” (red) and “Loss” (blue) and subsequently stratified by sex (filled vs unfilled boxes). (E) Kaplan-Meier plot showing overall survival trends for clear cell kidney cancers stratified by MAGEC3 (median cut) and sex.

Stratified at the median of tumor expression (corresponding to the 9.5th percentile of normal ovary expression), higher levels of MAGEC3 were not different from normal ovary (t-test p=0.368), while lower levels of MAGEC3 were significantly lower (p<0.001), reflecting loss of expression (Figure 1B). We subsequently named these tumor groups “Normal” and “Loss”. Normal MAGEC3 levels were associated with cases ascertained after 2006 (OR=5.86, 95%CI: 3.8-9.3, p<0.001) (Table 1).

In RCC patients, significantly lower MAGEC3 protein expression was identified in tumors (mean 146.6, median 145.0) compared to matched non-tumor samples (difference mean -22.13, median -23.33, paired t-test, p < 0.001) (Figure 1D). MAGEC3 protein levels were associated with black patients (OR=6.70, 95%CI:1.8-47.6, p=0.003) and also with non-clear cell (papillary) histology (OR=14.5, 95%CI: 4.1-99.5, p<0.001) (Table 2). Primary tissue expression in cases with synchronous metastases (N=73) was not different from non-metastatic cases (N=147) (148.8 vs 145.7, t-test, p=0.496), and metastatic expression was not different from their matched primaries (N=23, 146.6 vs 146.6, p=0.999).

MAGEC3 association with prognosis in epithelial ovarian cancer

In ovarian cancers with survival follow up (N=394), normal MAGEC3 levels were associated with complete response (CR) to maximal debulking surgery followed by platinum-based chemotherapy (CR mean 150.5 versus 138.9, two-sample t-test, p=0.008). Following initial complete response, women who were platinum sensitive (more than 9 months of progression-free survival after the end of platinum chemotherapy) continued to show higher MAGEC3 levels (platinum sensitive mean 149.8 versus 140.2, t-test p=0.033). The inverse was seen for progression-free survival (PFS) in all women with epithelial ovarian cancer (normal MAGEC3: 16.2 months median versus 24.1 months, log-rank p=0.002) (Table 1). Overall survival (OS) was significantly lower in women with normal MAGEC3 and optimal cytoreduction (R0, 64.7 months versus 118.6, p=0.050) but not in women with suboptimal cytoreduction (not R0, 33.9 versus 34.4, p=0.900).

The age at diagnosis, FIGO stage, grade, and R0 cytoreduction were univariately associated with prognosis in ovarian cancers (Table 3A). MAGEC3 level was independently associated with PFS (HR=1.41, 95%CI:1.14-1.75, p=0.002) (Figure 1C). Stepwise multivariate model selection by analysis of deviance led to a model that predicted progression-free survival using FIGO Stage (IIIC/IV, HR=3.14, score test p<0.001), histological grade (Poorly/Undifferentiated, HR=1.38, p=0.017), cytoreduction status (not R0, HR=1.45, p=0.022) and MAGEC3 level (Normal, HR=1.41, p=0.004) stratified on histological type and decade of life at diagnosis (global proportional hazards test p=0.32). Women were followed for an average of 52.6 months (maximum 244.4 months). These results were not affected by the year of ascertainment effect (PFS log-rank p=0.40, OS, p=0.20).

MAGEC3 sex-specific effect on overall survival in clear cell renal cell carcinoma

Recalling that MAGEC3 levels were lowest in clear cell tumors among kidney cancers, we focused on only clear cell tumors (N=162) to study a population with expected prognosis similar to that of epithelial ovarian cancers. Among clear cell RCC, overall survival was associated with age at resection, the presence of sarcomatoid features, and synchronous metastases at surgery (Table 3B). Additionally, MAGEC3 and sex were jointly associated with survival. With reference to women with MAGEC3 loss, neither women with normal MAGEC3 (HR=0.73, 0.31-1.72, p=0.50) nor men with MAGEC3 loss (HR=1.36, 0.73-2.55, p=0.30) had significantly different hazards. In contrast, men with normal levels had a significantly increased hazard (HR=2.34, 1.25-4.36, p=0.008) (Figure 1E).

Stepwise model selection led to a model predicting disease-specific overall survival including, in sequential order (Table 3B): histological presence of sarcomatoid cells (HR=1.49, final model p=0.017), age at surgery (+10 years, HR=1.77, p=0.002), size of tumor (+1 cm, HR=1.10, p=0.048), race (black, HR=6.02, p=0.030), and an interaction between sex and normal MAGEC3 level (p<0.001) stratified on the presence of metastasis at surgery (global proportional hazards test p=0.28). The MAGEC3 and sex interaction was interpreted as follows; with a reference level of women with MAGEC3 loss, women with normal MAGEC3 had a non-significant effect (HR=0.55, score test p=0.299), men with MAGEC3 loss were at increased risk (HR=2.07, p=0.056), and men with normal MAGEC3 were at a highly significant increased risk (HR=4.29, p=0.0003).

MAGEC3 is associated with CD8+ T cell infiltrated tumors

Given that ovarian cancers are highly immunogenic and that kidney tumors are responsive to immune checkpoint therapy, we considered the association between MAGEC3 and the count of CD8+ tumor infiltrating lymphocytes (TILs). We previously reported CD8+ TIL counts stained in ovarian cancer slides from these TMAs [12]. CD8+ TIL levels were positively correlated with MAGEC3 expression in ovarian tumors (Pearson’s r=0.176, p=0.011) (Figure S1A). We similarly stained and quantified the density of CD8+ TILs on the matching kidney TMAs, which had a similar trend (r=0.105, p=0.122) meriting further modeling. Adjusting for histology and the presence of metastatic disease, the correlation coefficient of the correctly scaled linear model was β=0.149 (score test p=0.0344) (Figure S1B).

Relationship with other CT antigens and seropositivity

We have previously reported on the expression of MAGE-A3, MAGE-A4, MAGE-A10, and PRAME [10, 13] in a subset of these ovarian cancers. Among all complete records (N=72), MAGEC3 was not strongly correlated with any MAGE-A antigens (A3, r=0.02, p=0.870; A4, r=0.16, p=0.185; A10, r=-0.06, p=0.604) (Figure S2A-C). A larger set of patients (N=195) had complete information on MAGEC3, PRAME (an autosomal antigen) [13] and NY-ESO-1 (gene CTAG1B at Xq28) expression and serology [11]. MAGEC3 was strongly correlated with PRAME expression (r=0.422, p<0.001, adjusted for year of diagnosis) (Figure S2D) and uncorrelated with NY-ESO-1 expression (r=0.101, p=0.157) (Figure S2E). Conversely, patients who were seropositive for NY-ESO-1 had upregulated expression of MAGEC3 (H-score β=11.2, p=0.0323) (Figure S2F). Logistic regression predicted NY-ESO-1 seropositivity was specific to MAGEC3 (+10 H-score units, OR=1.26, p=0.035) versus PRAME (OR=1.16, p=0.233).

MAGEC3 protein expression across multiple disease sites

FFPE blocks were available from cases submitted by Roswell Park to the Cancer Genome Atlas (TCGA) project from several tissue sites (N=180 tumors as a TMA). We stained these cores and analyzed their correlation with bulk mRNA sequencing data. Due to limited sample sizes within tissue sites, we considered a limited correlative analysis of protein expression in this set of patients (Table S1). As there is a known association between MAGEC3 and prognosis in ovarian and kidney cancers, we stratified common primary disease sites of the TCGA cohort by MAGEC3 expression. Only colorectal sites had an association with MAGEC3, displaying normal expression more frequently than MAGEC3 loss (p=0.04). When considering the whole cohort, CD8+ cell levels also have a notable association with MAGEC3 expression (CD8+ cell values associated with normal MAGEC3 expression) (p<0.001). Although race was found to be a factor associated with MAGEC3 levels in the kidney discovery cohort, this was not seen across this TCGA cohort (p=0.32).

MAGEC3 expression is associated with stress-related processes

MAGEC3-associated differentially expressed genes (DEGs, FDR≤0.5) adjusted for tissue site were more frequently upregulated across this TCGA subset (Figure 2A). We observed highly upregulated levels for the rest of the MAGEC locus (MAGEC1, and MAGEC2 on Xq27.2) as well as the neighboring MAGEA locus (MAGEA2, MAGEA3, MAGEA6, MAGEA10, and MAGEA12 on Xq28). Among these MAGE proteins, MAGEA2 and MAGEC2 interact with TRIM28 affecting the cell cycle among other processes [14, 15]. MAGEC3 also upregulates a stress related gene, WRNIP1 [16]; a DNA replication gene, MCM7 [17]; and cancer related genes, XPO5 [18] and ZSCAN16 [19]. In addition, MAGEC3 downregulates genes that are shown to induce tumor progression including APLMR [20, 21], FBLN5 [22], and FNDC1 [23, 24]. Gene set enrichment analysis (GSEA) showed that MAGEC3 significantly enriched E2F targets (NES=3.63, FDR<0.001), G2/M checkpoint (NES=3.20, FDR<0.001), and DNA repair (NES=2.28, FDR<0.001), whereas the epithelial to mesenchymal transition was downregulated (NES=-3.34, FDR<0.001) (Figure 2B). We also observed that differentially expressed genes were clustered in specific genomic locations such as 11p15, 11p23, 12p13, 17q21, and Xq28 (Figure 2C). These data suggest that MAGEC3 may induce a transcriptional response that involves the expression of stress-related genes that affect cell cycle and DNA repair, while downregulating genes involved in tumor progression.

• Download figure
• Open in new tab

Figure 2. MAGEC3 protein DEG and GSEA analyses and association with genomic instability.

(A) Volcano plot showing genes that are differentially expressed with continuous MAGEC3 protein expression. (B) Gene sets that are significantly associated with continuous MAGEC3 protein expression. (C) Ideogram showing where significantly expressed genes are located within the genome. Genes that were individually significant are listed above their locus and loci that are GSEA significant (FDR ≤ 0.05) are labelled. (D) Scatterplot showing the correlation between fraction genome altered and MAGEC3 protein levels in patients with PTEN, FAT4, BRAF, PTPRT, NF1, RB1, ATM, ATRX, IDH1, and TP63 mutations. (E) Comparison of DEGs between male and female patients.

MAGEC3 expression is correlated with genomic instability in patients with mutated oncogenes

Due to MAGEC3’s association with specific expression patterns in various genomic locations and upregulation of DNA repair processes, we investigated the association of MAGEC3 with genomic instability. We analyzed tumor mutation burden (TMB) data from 63 patients within the TCGA patient cohort stratified on the most mutated genes. A positive correlation between levels of MAGEC3 protein and fraction genome altered (Pearson’s r = 0.5, p<0.001) was observed in patients with PTEN, FAT4, BRAF, PTPRT, NF1, RB1, ATM, ATRX, IDH1, and TP63 mutations (Figure 2D). This association was negligible in patients without these oncogenic mutations (Pearson’s r = 0.03, p<0.001) (Figure S3).

MAGEC3 expression associates with different genes depending on sex

Stratifying by sex, we determined which DEGs were male or female specific or common to both groups (Figure 2E). Of particular interest was TSIX, a long non-coding RNA that inhibits XIST expression, thereby preventing X chromosome inactivation [25]. We observed a large log-fold change of TSIX expression in female patients with MAGEC3 protein expression indicative of a link between MAGEC3 protein expression and general X-chromosome inactivation. Another female specific DEG is the MHC class II gene HLA-DRA, which was specifically downregulated [26]. Male-specific DEGs include MCM7, a DNA replication licensing factor [17]; NKX2-2, an important biomarker for metastatic prostate cancers [27]; and NGFR, the canonical single-pass transmembrane receptor linked to melanoma [28] and squamous cell carcinoma [29].

Modeling MAGEC3 protein via mRNA

Much like other MAGE genes and X-linked cancer testis antigens, MAGEC3 had unusually low RNA counts despite measurable protein expression and demonstrated some linear correlation with protein expression (Pearson’s r=0.267, p=0.0003) likely due to a few high leverage points (Spearman’s ρ=-0.003, p=0.968). This observation suggested that, while some high expressing cases could be detected, RNA sequencing lacks the sensitivity to detect MAGEC3 mRNA levels.

To maximize the utility of mRNA datasets, we modeled protein expression of MAGEC3 using other mRNA measurements and divided the N=180 labeled TCGA set (Table S1) into training and validation datasets (2:1 ratio) where the TCGA clear cell kidney (KIRC) [30] and ovary (OV) [31] datasets were considered an unlabeled validation cohort. The final model selected by the 1 standard error (se) rule had a training mean square error (MSE) of 455.5 (Figure 3A) and a leave out one test MSE of 637.9 (N=120, se: 110.11). The fully withheld validation set MSE was 885.3 (N=60). Correlation between true MAGEC3 protein scores and predicted scores was 0.69 (p<0.001) in the model building cohort and 0.51 (p<0.001) in the validation cohort (Figure 3B-C). Similar error rates and correlations, such as those identified within these cohorts, imply a well-fit model with predictive modeling capabilities. The fitted model selected mRNA expression of AMBN, CCDC77, RNF175, C7orf51, MCART3P, CBLL1, and KIAA1429 as positively correlated with MAGEC3 protein and RPS6KA2, LYVE1, CCDC80, CX3CR1, EMP1, CDSN, and ANGPT4 as negatively correlated with MAGEC3 protein.

• Download figure
• Open in new tab

Figure 3. RNA-based predictive model for MAGEC3 expression.

(A) Leave out one (LOO) cross-validation error for tuning parameter selection. Fitted values in the (B) training set and (C) withheld test set. Survival estimate validation in the independent pan-TCGA dataset for (D) ovarian cancers, (E) kidney cancer males, and (F) kidney cancer females.

MAGEC3 protein predictor in Pan-TCGA study

We applied the MAGEC3 protein level predictor to an independent set of cases sequenced and processed for TCGA’s pan-cancer analysis (N=282 ovary, N=606 clear cell RCC) which did not have measured MAGEC3 protein levels. Analyses performed on our discovery cohorts were also performed on these predicted MAGEC3 scores. In ovarian cancers, progression-free survival had a significant negative relationship with normal MAGEC3 scores (HR=1.35 p=0.048) which persisted after accounting for age, stage, and initial response to chemotherapy (HR =1.74 p=0.002). When split at the median, higher MAGEC3 cases had a median survival of 36.9 months (34.8-43.7) versus 51.3 months (95% interval: 45.0-62.1) (Figure 3D). In clear cell RCCs, men with high MAGEC3 levels had significantly poorer overall survival (log-rank p=0.003) (Figure 3E). Under proportional hazards, the effect (HR=1.67, p=0.004) was of similar magnitude in women (HR=1.62, p=0.045) (Figure 3F).

Patient cohorts and TMA construction

Three cohorts of patients were used in this study: (1) a “discovery cohort” of ovarian and kidney cancer patients from Roswell Park, (2) a “model-building” cohort of mutually exclusive patients from Roswell Park that had been accrued and molecularly characterized through the TCGA project and (3) a second validation cohort of TCGA patients from other institutions without protein data.

With respect to the first cohort, following review by the Roswell Park Institutional Review Board, we obtained clinical data and archival samples for: (a) women with a primary diagnosis of ovarian cancer treated by maximal debulking surgery and first line platinum-based chemotherapy and (b) patients with a primary diagnosis of renal cell carcinoma undergoing nephrectomy at the institution. We required that patients in both cohorts had available tumor blocks taken prior to the initiation of systemic therapy. In total, there were N=220 RCC patients and N=411 ovarian cancer patients (high-grade serous ovarian cancer, N=333) for analysis. Tissue blocks were organized into tissue microarrays as previously described [10, 46] and stained for MAGEC3 protein expression. TMAs contained reference cores for normalization across different blocks.

Antibody validation

The MAGEC3 antibody was vetted by the Human Protein Atlas [45] (HPA052067) following the guidelines of the International Working Group for Antibody Validation [47] and was scored as approved for IHC and supported for antigen specificity. The antigen appears to bind both dominant isoforms of MAGEC3 (ESNP000000386566 and ENSP00000440444) outside of the functional MAGE homology domain.

Immunohistochemistry

Formalin-fixed paraffin sections were cut at 4µm, placed on charged slides, and dried at 60°C for one hour. Slides were cooled to room temperature and added to the Dako Omnis autostainer, where they were deparaffinized with Clearify (American Mastertech; catalog# CACLEGAL) and rinsed in water. Flex TRS High pH (Dako; catalog# GV804) was used for target retrieval for 60 minutes. Slides were incubated with MAGEC3 antibody (Sigma Aldrich; catalog# HPA052067) for 50 minutes at 1:50 dilution. Flex Rabbit Linker (Dako; catalog# GV80911-2) was applied for 10 minutes followed by Flex /HRP polymer (Dako; catalog# DM843) which was applied for 20 minutes followed by DAB (Diaminobenzidine) (Dako; catalog# K3468) for 5 minutes for visualization. Slides were counterstained with Hematoxylin for 8 minutes then put into water. After removing slides from the Omnis they are dehydrated, cleared, and cover slipped. Cores were averaged by patient and checked for outliers and slides were normalized to common internal control spots included in the design of the TMA. MAGEC3 was quantified using an IHC Profiler [48]. CD8 (Dako #M7103) was stained as previously described [12]. MAGEA3, MAGEA4 and MAGEA10 were stained as previously described [10] and scored as positive or negative. PRAME was stained and quantitatively scored as previously described [13]. After staining, slide-specific staining intensity and rank and file associations were tested to identify any slide/batch or spatial effects.

TCGA study data

We downloaded clinical and expression data for the TCGA Pan-Cancer (PANCAN) cohort directly from UCSC Xena browser as batch effect normalized mRNA data (log2 RSEM, upper-quartile normalized [49]). Modifications made to the data after download include the removal of unknown or hypothetical transcripts in the Entrez database and the setting of NA counts to 0. Corresponding clinical information for the expression data were obtained from PANCAN’s curated clinical data [50].

Statistical analysis

Statistical tests are described as they are used in text. In general, Student’s t-test, chi-square test and log-rank test were employed as appropriate for continuous, categorical or survival time data. All statistical tests were two-sided and p<0.05 was considered significant throughout. Survival analysis by Cox’s proportional hazards model used analysis of deviance-based model selection and employed residual tests for non-proportionality [51]. Overall survival was defined as a cancer associated death event censored by end of follow up. For ovarian cancers, progression-free survival was defined as doubling of baseline CA125 confirmed with CT scan or the initiation of relapse chemotherapy regimens as defined in [52]. All analyses were performed in the R statistical programming language version 3.6.1.

LASSO model building

Within the model-building cohort, genes with consistent univariate association were screened for MAGEC3 association and the top 1000 features with univariate association with MAGEC3 [53] were selected for multivariate model building. We used a LASSO-penalized linear model objective to select and estimate coefficients and selected the tuning parameter by unbiased leave-out-one (LOO) cross validation via R/glmnet [54]. Suboptimal local solutions were ruled out by multiple randomized partitions of the model/validation cohorts.

DEG calling was performed in R/limma using the moderated t-statistics and FDR control [55]. The linear model included the TCGA study as a covariate. Gene set analysis was performed using the R/fgsea package [56]. Annotations were maintained via R/org.Hs.eg.db and MSigDb’s Hallmark pathway set. Pathways with fewer than 15 represented genes were removed from consideration.